Chemical Process Safety Fundamentals With Applications,2ND ED

Uploaded by: Ebooks Chemical Engineering (https://www.facebook.com/pages/Ebooks-Chemical-Engineering/238197077030)

For More Books, softwares & tutorials Related to Chemical Engineering

Join Us

@facebook: https://www.facebook.com/AllAboutChemcalEngineering

@facebook: https://www.facebook.com/groups/10436265147/

@facebook: https://www.facebook.com/pages/Ebooks-Chemical-

Engineering/238197077030

ADMIN:

I.W

Chemical Process Safety

ISBN 0-13-OZ817b-5

9 780130 181763

PRENTICE HALL INTERNATIONAL SERIES IN THE PHYSICAL AND CHEMICAL ENGINEERING SCIENCES

NEAL R. AMUNDSON, SERIES EDITOR, University of Houston

ANDREAS ACRIVOS, Stanford University JOHN DAHLER, University of Minnesota

H. SCOTT FOGLER, University of Michigan THOMAS J. HANRATTY, University of Illinois JOHN M . PRAUSNITZ, University of California

L. E. SCRIVEN, University of Minnesota

BALZHISER, SAMUELS, AND ELIASSEN Chemical Engineering Thermodynamics BEQUETTE Process Control: Modeling, Design and Simulation BEQUETTE Process Dynamics BIEGLER, GROSSMAN, AND WESTERBERG Systematic Methods of Chemical Process

Design BROSILOW AND JOSEPH Techniques of Model-Based Control CROWL AND LOUVAR Chemical Process Safety: Fundamentals with Applications,

2nd edition CONSTANTINIDES AND MOSTOUFI Numerical Methods for Chemical Engineers

with MATLAB Applications CUTLIP AND SHACHAM Problem Solving in Chemical Engineering with Numerical

Methods DENN Process Fluid Mechanics DOYLE Process Control Modules: A Software Laboratory for Control Design ELLIOT AND LIRA Introductory Chemical Engineering Thermodynamics FOGLER Elements of Chemical Reaction Engineering, 3rd edition HIMMELBLAU Basic Principles and Calculations in Chemical Engineering, 6th edition HINES AND MADDOX Mass Transfer KYLE Chemical and Process Thermodynamics, 3rd edition PRAUSNITZ, LICHTENTHALER, AND DE AZEVEDO Molecular Thermodynamics

of Fluid-Phase Equilibria, 3rd edition PRENTICE Electrochemical Engineering Principles SHULER AND KARGI Bioprocess Engineering, 2nd edition STEPHANOPOULOS Chemical Process Control TESTER AND MODELL Thennodynainics and Its Applications, 3rd edition TURTON, BAILIE, WHITING, AND SHAEIWITZ Analysis, Synthesis and Design

of Chemical Processes WILKES Fluid Mechanics for Chemical Engineering

Prentice Hall International Series in the Physical and Chemical Engineering Sciences

Chemical Process Safety Fundamentals with Applications Second Edition

Daniel A. Crow1 Michigan Technological University

Joseph F. Louvar Wayne State University

Prentice Hall PTR Upper Saddle River, New Jersey 07458

www.phptr.com

Library of Congress Cataloging-in-Publication data Crowl, Daniel A.

Chemical process safety : fundamentals with applications I Daniel A. Crowl, Joseph F. Louvar. - 2nd ed.

p. cm. - (Prentice Hall international series in the physical and chemical engineering sciences) Includes bibliographical references and index. ISBN 0-13-018176-5 1. Chemical plants -Safety measures. I. Louvar, Joseph F. 11. Title. 111. Series.

EditoriallProduction Supervision: G&S Typesetters, Inc. Acquisitions Editor: Bernard Goodwin Marketing Manager: Dan DePasquale Cover Design Direction: Jerry Votta Art Director: Gail Cocker-Bogusz Manufacturing Manager: Alexis R. Heydt-Long Editorial Assistant: Michelle Vincenti Project Coordinator: Anne R. Garcia

O 2002 by Prentice Hall PTR Prentice Hall, Inc. Upper Saddle River, NJ 07458

Prentice Hall books are widely used by corporations and government agencies for training, marketing, and resale.

The publisher offers discounts on this book when ordered in bulk quantities. For more information, contact Corporate Sales Department, phone: 800-382-3419; fax: 201-236-7141; email: [email protected] Or write: Prentice Hall PTR

Corporate Sales Department One Lake Street Upper Saddle River, NJ 07458

All rights reserved. No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher.

Printed in the United States of America 1 0 9 8 7 6 5 4 3 2 1

ISBN: 0-13-018176-5

Pearson Education Ltd. Pearson Education Australia PTY, Ltd. Pearson Education Singapore, Pte. Ltd. Pearson Education North Asia Ltd. Pearson Education Canada, Ltd. Pearson Educaci6n de Mexico, S.A. de C.V. Pearson Education - Japan Pearson Education Malaysia, Pte. Ltd. Pearson Education, Upper Saddle New Jersey

Contents

Preface xiii Nomenclature xv

1 Introduction 1 Safety Programs 2 Engineering Ethics 4 Accident and Loss Statistics 4 Acceptable Risk 12 Public Perceptions 14 The Nature of the Accident Process 15 Inherent Safety 20 Four Significant Disasters 23

Flixborough, England 23 Bhopal, India 25 Seveso, Italy 26 Pasadena, Texas 27

Suggested Reading 29 Problems 30

2 Toxicology 35 2-1 How Toxicants Enter Biological Organisms 36

Gastrointestinal Tract 37 Skin 37 Respiratory System 38

2-2 How Toxicants Are Eliminated from Biological Organisms 39 2-3 Effects of Toxicants on Biological Organisms 40 2-4 Toxicological Studies 41 2-5 Dose versus Response 42

vi Contents

2-6 Models for Dose and Response Curves 48 2-7 Relative Toxicity 54 2-8 Threshold Limit Values 54


3 Industrial Hygiene 63 3-1 Government Regulations 64

Laws and Regulations 64 Creating a Law 64 Creating a Regulation 64 OSHA: Process Safety Management 68 EPA: Risk Management Plan 71

3-2 Industrial Hygiene: Identification 74 Material Safety Data Sheets 74

3-3 Industrial Hygiene: Evaluation 78 Evaluating Exposures to Volatile Toxicants by Monitoring 79 Evaluation of Worker Exposures to Dusts 83 Evaluating Worker Exposures to Noise 84 Estimating Worker Exposures to Toxic Vapors 85

3-4 Industrial Hygiene: Control 94 Respirators 96 Ventilation 97


Source Models 109 4-1 Introduction to Source Models 109 4-2 Flow of Liquid through a Hole 112 4-3 Flow of Liquid through a Hole in a Tank 116 4-4 Flow of Liquids through Pipes 121

2-K Method 124 4-5 Flow of Vapor through Holes 130 4-6 Flow of Gases through Pipes 136

Adiabatic Flows 136 Isothermal Flows 143

4-7 Flashing Liquids 151 4-8 Liquid Pool Evaporation or Boiling 157 4-9 Realistic and Worst-Case Releases 159 4-10 Conservative Analysis 159


Contents vii

5 Toxic Release and Dispersion Models 171 5-1 Parameters Affecting Dispersion 172 5-2 Neutrally Buoyant Dispersion Models 176

Case 1: Steady-State Continuous Point Release with No Wind 180 Case 2: Puff with No Wind 181 Case 3: Non-Steady-State Continuous Point Release with No Wind 182 Case 4: Steady-State Continuous Point Source Release with Wind 183 Case 5: Puff with No Wind and Eddy Diffusivity Is a Function of Direction

183 Case 6: Steady-State Continuous Point Source Release with Wind and Eddy

Diffusivity Is a Function of Direction 184 Case 7: Puff with Wind 184 Case 8: Puff with No Wind and with Source on Ground 185 Case 9: Steady-State Plume with Source on Ground 185 Case 10: Continuous Steady-State Source with Source at Height Hr above

the Ground 186 Pasquill-Gifford Model 186 Case 11: Puff with Instantaneous Point Source at Ground Level, Coordi-

nates Fixed at Release Point, Constant Wind Only in x Direction with Constant Velocity u 190

Case 12: Plume with Continuous Steady-State Source at Ground Level and Wind Moving in x Direction at Constant Velocity u 191

Case 13: Plume with Continuous Steady-State Source at Height Hr above Ground Level and Wind Moving in x Direction at Constant Velocity u 192

Case 14: Puff with Instantaneous Point Source at Height Hr above Ground Level and a Coordinate System on the Ground That Moves with the Puff 193

Case 15: Puff with Instantaneous Point Source at Height Hr above Ground Level and a Coordinate System Fixed on the Ground at the Release Point 194

Worst-Case Conditions 194 Limitations to Pasquill-Gifford Dispersion Modeling 194

5-3 Dense Gas Dispersion 195 5-4 Toxic Effect Criteria 199 5-5 Effect of Release Momentum and Buoyancy 212 5-6 Release Mitigation 213


6 Fires and Explosions 225 6-1 The Fire Triangle 225 6-2 Distinction between Fires and Explosions 227

- - - - -

viii Contents

6-3 Definitions 227 6-4 Flammability Characteristics of Liquids and Vapors 229

Liquids 230 Gases and Vapors 233 Vapor Mixtures 233 Flammability Limit Dependence on Temperature 235 Flammability Limit Dependence on Pressure 236 Estimating Flammability Limits 236

6-5 Limiting Oxygen Concentration and Inerting 238 6-6 Flammability Diagram 240 6-7 Ignition Energy 248 6-8 Autoignition 249 6-9 Auto-Oxidation 249 6-10 Adiabatic Compression 249 6-11 Ignition Sources 251 6-12 Sprays and Mists 252 6-13 Explosions 252

Detonation and Deflagration 253 Confined Explosions 255 Blast Damage Resulting from Overpressure 265 TNT Equivalency 269 TNO Multi-Energy Method 271 Energy of Chemical Explosions 274 Energy of Mechanical Explosions 276 Missile Damage 279 Blast Damage to People 279 Vapor Cloud Explosions 281 Boiling-Liquid Expanding-Vapor Explosions 282


7 Designs to Prevent Fires and Explosions 291 7-1 Inerting 292

Vacuum Purging 292 Pressure Purging 295 Combined Pressure-Vacuum Purging 297 Vacuum and Pressure Purging with Impure Nitrogen 298 Advantages and Disadvantages of the Various Pressure and Vacuum Tnert-

ing Procedures 299 Sweep-Through Purging 299 Siphon Purging 301 Using the Flammability Diagram To Avoid Flammable Atmospheres 301

7-2 Static Electricity 307 Fundamentals of Static Charge 307

--

Contents ix

Charge Accumulation 308 Electrostatic Discharges 309 Energy from Electrostatic Discharges 311 Energy of Electrostatic Ignition Sources 312 Streaming Current 313 Electrostatic Voltage Drops 316 Energy of Charged Capacitors 316 Capacitance of a Body 321 Balance of Charges 324

7-3 Controlling Static Electricity 330 General Design Methods To Prevent Electrostatic Ignitions 333 Relaxation 332 Bonding and Grounding 332 Dip Pipes 333 Increasing Conductivity with Additives 336 Handling Solids without Flammable Vapors 337 Handling Solids with Flammable Vapors 337

7-4 Explosion-Proof Equipment and Instruments 337 Explosion-Proof Housings 339 Area and Material Classification 339 Design of an XP Area 340

7-5 Ventilation 340 Open-Air Plants 340 Plants Inside Buildings 341

7-6 Sprinkler Systems 343 7-7 Miscellaneous Designs for Preventing Fires and Explosions 347


8 Introduction to Reliefs 353 8-1 Relief Concepts 354 8-2 Definitions 356 8-3 Location of Reliefs 357 8-4 Relief Types 360 8-5 Relief Scenarios 364 8-6 Data for Sizing Reliefs 365 8-7 Relief Systems 368

Relief Installation Practices 368 Relief Design Considerations 368 Horizontal Knockout Drum 371 Flares 375 Scrubbers 376 Condensers 376


x Contents

9 Relief Sizing 383 9-1 Conventional Spring-Operated Reliefs in Liquid Service 384 9-2 Conventional Spring-Operated Reliefs in Vapor or Gas Service 389 9-3 Rupture Disc Reliefs in Liquid Service 394 9-4 Rupture Disc Reliefs in Vapor or Gas Service 394 9-5 Two-Phase Flow during Runaway Reaction Relief 395

Simplified Nomograph Method 401 9-6 Deflagration Venting for Dust and Vapor Explosions 404

Vents for Low-Pressure Structures 406 Vents for High-Pressure Structures 408

9-7 Venting for Fires External to Process Vessels 411 9-8 Reliefs for Thermal Expansion of Process Fluids 415


10 Hazards Identification 429 10-1 Process Hazards Checklists 432 10-2 Hazards Surveys 432 10-3 Hazards and Operability Studies 448 10-4 Safety Reviews 454 10-5 Other Methods 459


11 Risk Assessment 471 11-1 Review of Probability Theory 472

Interactions between Process Units 474 Revealed and Unrevealed Failures 480 Probability of Coincidence 484 Redundancy 486 Common Mode Failures 486

11-2 Event Trees 486 11-3 Fault Trees 491

Determining the Minimal Cut Sets 494 Quantitative Calculations Using the Fault Tree 497 Advantages and Disadvantages of Fault Trees 498 Relationship between Fault Trees and Event Trees 498

11-4 QRA and LOPA 499 Quantitative Risk Analysis 499 Layer of Protection Analysis 500 Consequence 503 Frequency 503


Contents xi

Accident Investigations 515 12-1 Learning from Accidents 515 12-2 Layered Investigations 516 12-3 Investigation Process 518 12-4 Investigation Summary 519 12-5 Aids for Diagnosis 521

Fires 522 Explosions 522 Sources of Ignition in Vessels 523 Pressure Effects 523 Medical Evidence 525 Miscellaneous Aids to Diagnosis 525

12-6 Aids for Recommendations 528 Control Plant Modifications 528 User-Friendly Designs 529 Block Valves 529 Double Block and Bleed 530 Preventive Maintenance 530 Analyzers 531


13 Case Histories 535 13-1 Static Electricity 536

Tank Car Loading Explosion 536 Explosion in a Centrifuge 536 Duct System Explosion 537 Conductor in a Solids Storage Bin 537 Pigment and Filter 536 Pipefitter's Helper 536 Lessons Learned 536

13-2 Chemical Reactivity 540 Bottle of Isopropyl Ether 540 Nitrobenzene Sulfonic Acid Decomposition 540 Organic Oxidation 541 Lessons Learned 541

13-3 System Designs 546 Ethylene Oxide Explosion 546 Ethylene Explosion 546 Butadiene Explosion 546 Light Hydrocarbon Explosion 547 Pump Vibration 547 Pump Failure 547 Ethylene Explosion (1) 548

xii Contents

Ethylene Explosion (2) 548 Ethylene Oxide Explosion 548 Lessons Learned 549

13-4 Procedures 551 Leak Testing a Vessel 552 Man Working in Vessel 552 Vinyl Chloride Explosion 552 Dangerous Water Expansion 553 Phenol-Formaldehyde Runaway Reaction 553 Conditions and Secondary Reaction Cause Explosion 554 Fuel-Blending Tank Explosion 555 Lessons Learned 556

13-5 Conclusion 556 Suggested Reading 557 Problems 557

Appendix A: Unit Conversion Constants 561

Appendix B: Flammability Data for Selected Hydrocarbons 565

Appendix C: Detailed Equations for Flammability Diagrams 571 Equations Useful for Placing Vessels into and out of Service 576

Appendix D: Formal Safety Review Report for Example 10-4 581

Appendix E: Saturation Vapor Pressure Data 591

Preface

T his second edition of Chemical Process Safety is designed to enhance the process of teaching and applying the fundamentals of chemical process safety. It is appropriate for an industrial reference, a senior-level undergraduate course, or a graduate course in chemical process safety. It can be used by anyone interested in improving chemical process safety, including chemical and mechanical engineers and chemists. More material is presented than can be accommodated in a 3-credit course, providing instructors with the opportunity to emphasize their topics of interest.

The primary objective of this textbook is to encapsulate the important technical fundamentals of chemical process safety. The emphasis on the fundamentals will help the student and practicing scientist to understand the concepts and apply them accordingly. This application requires a significant quantity of fundamental knowledge and technology.

The second edition has been rewritten to include new process safety technology and new references that have appeared since the first edition was published in 1990. It also includes our combined experiences of teaching process safety in both industry and academia during the past 10 years.

Significant modifications were made to the following topics: dispersion modeling, source modeling, flammability characterization, explosion venting, fundamentals of electrostatics, and case histories. This new edition also includes selected materials from the latest AICHE Center for Chemical Process Safety (CCPS) books and is now an excellent introduction to the CCPS library.

This second edition also includes more problems (now 30 per chapter). A complete set of problem solutions is available to instructors using the book in their curriculum. These changes fulfill the requests of many professors who have used this textbook.

We continue to believe that a textbook on safety is possible only with both industrial and academic inputs. The industrial input ensures that the material is industrially relevant. The

xiv Preface

academic input ensures that the material is presented on a fundamental basis to help professors and students understand the concepts. Although the authors are (now) both from universities, one has over 30 years of relevant experience in industry (J. F. L.) and the other (D. A. C.) has accumulated significant industrial experience since the writing of the first edition.

Since the first edition was published, many universities have developed courses or course content in chemical process safety. This new emphasis on process safety is the result of the positive influences from industry and the Accreditation Board for Engineering and Technology (ABET). Based on faculty feedback, this textbook is an excellent application of the fundamental topics that are taught in the first three years of the undergraduate education.

Although professors normally have little background in chemical process safety, they have found that the concepts in this text and the accompanying problems and solutions are easy to learn and teach. Professors have also found that industrial employees are enthusiastic and willing to give specific lectures on safety to enhance their courses.

This textbook is designed for a dedicated course in chemical process safety. However, we continue to believe that chemical process safety should be part of every undergraduate and graduate course in chemistry and chemical and mechanical engineering, just as it is a part of all the industrial experiences. This text is an excellent reference for these courses. This textbook can also be used as a reference for a design course.

Some will remark that our presentation is not complete or that some details are missing. The purpose of this book, however, is not to be complete but to provide a starting point for those who wish to learn about this important area. This book, for example, has a companion text titled Health and Environmental Risk Analysis that extends the topics relevant to risk analysis.

We thank many of our friends who continue to teach us the fundamentals of chemical process safety. Those who have been especially helpful include G. Boicourt and J. Wehman of the BASF Corporation; W. Howard and S. Grossel, who have extensive industrial experience and are now consultants; B. Powers from Dow Chemical Company; D. Hendershot from Rohm and Haas; R. Welker of the University of Arkansas; R. Willey of Northeastern University; and R. Darby of Texas A&M University.

We also continue to acknowledge and thank all the members of the Undergraduate Ed- ucation Committee of the Center for Chemical Process Safety and the Safety and Loss Pre- vention Committee of the American Institute of Chemical Engineers. We are honored to be members of both committees. The members of these committees are the experts in safety; their enthusiasm and knowledge have been truly educational and a key inspiration to the development of this text.

Finally, we continue to acknowledge our families, who provided patience, understanding, and encouragement throughout the writing of the first and second editions.

We hope that this textbook helps prevent chemical plant and university accidents and contributes to a much safer future.

Daniel A. Crowl and Joseph E: Louvar

Nomenclature

Do D", Dtid

Ea ERPG

velocity of sound (lengthltime) area (length2) or Helmholtz free energy (energy); or process component availability tank cross sectional area (length2) change in Helmoltz free energy (energylmole) mass concentration (masslvolume) or capacitance (Farads) discharge coefficients (unitless) or concentration at a specified time (mass/volume) concentration of dense gas (volume fraction) heat capacity at constant pressure (energylmass deg) heat capacity at constant volume (energylmass deg) concentration in parts per million by volume deflagration vent constant (pressure1'*) average or mean mass concentration (mass/volume) diameter (length) particle diameter (length) diameter of flare stack (length) diffusion coefficient (arealtime characteristic source dimension for continuous releases of dense gases (length) characteristic source dimension for instantaneous releases of dense gas (length) reference diffusion coefficient (arealtime) molecular diffusivity (area /time) total integrated dose due to a passing puff of vapor (mass timelvolume) activation energy (energylmole) emergency response planning guideline (see Table 5-6)

xvi Nomenclature

EEGL f f ( t) f v F FAR FEV FVC g gc go G GT AG h h~ h"L hs H Hf HI AH AHC AH, AH" I ID IDLH 10 Is ISOC j J k kl, k2 ks K Kb Kf Ki, Krn KG Kj KP Kst K"

emergency exposure guidance levels (see section 5.4) Fanning friction factor (unitless) or frequency (lltime) failure density function mass fraction of vapor (unitless) frictional fluid flow loss term (energy mass) or force or environment factor fatal accident rate (fatalitiesIlO8 hours) forced expired volume (literslsec) forced vital capacity (liters) gravitational acceleration (lengthltime2) gravitational constant initial cloud buoyancy factor (lengthltime2) Gibbs free energy (energylmole) or mass flux (masslarea time) mass flux during relief (masstarea time) change in Gibbs free energy (energylmole) specific enthalpy (energylmass) fluid level above leak in tank (length) initial fluid level above leak in tank (length) leak height above ground level (length) enthalpy (energylrnole) or height (length) flare height (length) effective release height in plume model (length) change in enthalpy (energylrnole) heat of combustion (energylmass) release height correction given by Equation 5-64 enthalpy of vaporization (energylmass) sound intensity (decibels) pipe internal diameter (length) immediately dangerous to life and health (see section 5.4) reference sound intensity (decibels) streaming current (amps) in-service oxygen concentration (volume percent oxygen) number of inerting purge cycles (unitless) electrical work (energy) non-ideal mixing factor for ventilation (unitless) constants in probit a equations thermal conductivity of soil (energyllength time deg) mass transfer coefficient (lengthltime) backpressure correction for relief sizing (unitless) excess head loss for fluid flow (dimensionless) constants in excess head loss, given by Equation 4-38 explosion constant for vapors (length pressureltime) eddy diffusivity in x, y or z direction (areattime) overpressure correction for relief sizing (unitless) explosion constant for dusts (length pressureltime) viscosity correction for relief sizing (unitless)

Nomenclature xvii

KO K* L LEL LFL = LEL LOC m

"no

~ T N T

mv M Mo Ma MOC, MSOC MTBC MTBF n

OSFC P Pd Ps P P b PEL PFD PP Pmax

f's psat

4 4f qg 4s

Q Q", Q; Qv r

R -

R Rd RHI rf RP

reference mass transfer coefficient (lengthltime) constant eddy diffusivity (arealtime) length lower explosion limit (volume %) lower flammability limit (volume %) limiting oxygen concentration (volume percent oxygen) mass

total mass contained in reactor vessel (mass) mass of TNT mass of vapor molecular weight (masslmole) reference molecular weight (masslmole) Mach number (unitless) See LOC mean time between coincidence (time) mean time between failure (time) number of moles out of service fuel concentration (volume percent fuel) partial pressure (forcelarea) number of dangerous process episodes scaled overpressure for explosions (unitless) total pressure or probability backpressure for relief sizing (psig) permissable exposure level (see section 5.4) probability of failure on demand gauge pressure (forcelarea) maximum pressure for relief sizing (psig) set pressure for relief sizing (psig) saturation vapor pressure heat (energylmass) or heat intensity (energylarea time) heat intensity of flare (energyltime area) heat flux from ground (energylarea time) specific energy release rate at set pressure during reactor relief (energylmass) heat (energy) or electrical charge (coulombs) mass discharge rate (massltime) instantaneous mass release (mass) ventilation rate (volumeltime) radius (length) electrical resistance (ohms) or reliability Sachs scaled distance, defined by equation 6-25 (unitless) release duration for heavy gas releases (time) reaction hazard index defined by Equation 13-1 vessel filling rate (time-') ideal gas constant (pressure volume/mole deg)

xviii Nomenclature

Re S s m

SPEGL t

td

te tl, t" tw

At" T Td Ti TLV T m

TWA TXD U

U d -

U

(4 U

UEL UFL = UEL v

Vf

v g

vfs v v c

W we

ws X

Xf Y Y YG Z

Reynolds number (unitless) entropy (energylmole deg) or stress (forcelarea) material strength (forcelarea) short term public exposure guideline (see section 5.4) time positive phase duration of a blast (time) emptying time time to form a puff of vapor vessel wall thickness (length) worker shift time venting time for reactor relief temperature (deg) material decomposition temperature (deg) time interval threshold limit value (ppm or mg/m3 by volume) maximum temperature during reactor relief (deg) saturation temperature at set pressure during reactor relief (deg) time weighted average (ppm or mg/m3 by volume) toxic dispersion method (see section 5.4) velocity (lengthltime) dropout velocity of a particle (lengthltime) average velocity (lengthltime) mean or average velocity (lengthhime) internal energy (energylmole) or overall heat transfer coefficient (energylarea time) or process component unavailability upper explosion limit (volume %) upper flammability limit (volume %) specific volume (volumelmass) specific volume of liquid (volumelmass) specific volume of vapor (volumelmass) specific volume change with liquid vaporization (volumelmass) total volume or electrical potential (volts) container volume width (length) expansion work (energy) shaft work (energy) mole fraction or Cartesian coordinate (length) distance from flare at grade (length) mole fraction of vapor (unitless) or Cartesian coordinate (length) probit variable (unitless) gas expansion factor (unitless) height above datum (length) or Cartesian coordinate (length) or com- pressibility (unitless) scaled distance for explosions (lengthlma~sl'~)

Nomenclature xix

Greek Letters velocity correction factor (unitless) or thermal diffusivity (arealtime) thermal expansion coefficient (deg-') double layer thickness (length) pipe roughness (length) or emissivity (unitless) relative dielectric constant (unitless) permittivity constant for free space (charge2/force length2) explosion efficiency (unitless) nonideal filling factor (unitless) heat capacity ratio (unitless) conductivity (mholcm) function defined by Equation 9-6 frequency of dangerous episodes average frequency of dangerous episodes viscosity (mass/length/time) or mean value or failure rate (faultsltime) vapor viscosity (mass/length/time) overall discharge coefficient used in Equation 9-15 (unitless) density (mass/volume) liquid density (mass/volume) reference density for specific gravity (mass/volume) vapor density (mass/volume) standard deviation (unitless) dispersion coefficient (length) relaxation time inspection period for unrevealed failures operation period for a process component period required to repair a component period of unavailability for unrevealed failures zeta potential (volts)

Subscripts Superscripts ambient combustion formation or liquid vapor or gas higher pressure initiating event purges lower pressure maximum set pressure initial or reference

0 standard I stochastic or random variable

Introduction

I n 1987, Robert M. Solow, an economist at the Massa- chusetts Institute of Technology, received the Nobel Prize in economics for his work in determining the sources of economic growth. Professor Solow concluded that the bulk of an econ- omy's growth is the result of technological advances.

It is reasonable to conclude that the growth of an industry is also dependent on technological advances. This is especially true in the chemical industry, which is entering an era of more complex processes: higher pressure, more reactive chemicals, and exotic chemistry.

More complex processes require more complex safety technology. Many industrialists even believe that the development and application of safety technology is actually a constraint on the growth of the chemical industry.

As chemical process technology becomes more complex, chemical engineers will need a more detailed and fundamental understanding of safety. H. H. Fawcett said, "To know is to sur- vive and to ignore fundamentals is to court disaster." l This book sets out the fundamentals of chemical process safety.

Since 1950, significant technological advances have been made in chemical process safety. Today, safety is equal in importance to production and has developed into a scientific discipline that includes many highly technical and complex theories and practices. Examples of the technology of safety include

hydrodynamic models representing two-phase flow through a vessel relief, dispersion models representing the spread of toxic vapor through a plant after a release, and

'H. H. Fawcett and W. S. Wood, Safety andAccident Prevention in Chemical Operations, 2d ed. (New York: Wiley, 1982), p. 1.

-

2 Chapter I Introduction

mathematical techniques to determine the various ways that processes can fail and the probability of failure.

Recent advances in chemical plant safety emphasize the use of appropriate technological tools to provide information for making safety decisions with respect to plant design and operation.

The word "safety" used to mean the older strategy of accident prevention through the use of hard hats, safety shoes, and a variety of rules and regulations. The main emphasis was on worker safety. Much more recently, "safety" has been replaced by "loss prevention." This term includes hazard identification, technical evaluation, and the design of new engineering features to prevent loss. The subject of this text is loss prevention, but for convenience, the words "safety" and "loss prevention" will be used synonymously throughout.

Safety, hazard, and risk are frequently-used terms in chemical process safety. Their definitions are

Safety or loss prevention: the prevention of accidents through the use of appropriate tech- nologies to identify the hazards of a chemical plant and eliminate them before an accident occurs.

Hazard: a chemical or physical condition that has the potential to cause damage to people, property, or the environment. Risk: a measure of human injury, environmental damage, or economic loss in terms of both the incident likelihood and the magnitude of the loss or injury.

Chemical plants contain a large variety of hazards. First, there are the usual mechanical hazards that cause worker injuries from tripping, falling, or moving equipment. Second, there are chemical hazards. These include fire and explosion hazards, reactivity hazards, and toxic hazards.

As will be shown later, chemical plants are the safest of all manufacturing facilities. How- ever, the potential always exists for an accident of catastrophic proportions. Despite substantial safety programs by the chemical industry, headlines of the type shown in Figure 1-1 continue to appear in the newspapers.

1-1 Safety Programs A successful safety program requires several ingredients, as shown in Figure 1-2. These ingredients are

System Attitude Fundamentals Experience Time You

1-1 Safety Programs 3

Figure 1-1 Headlines are indicative of the public's concern over chemical safety.

First, the program needs a system (1) to record what needs to be done to have an outstanding safety program, (2) to do what needs to be done, and (3) to record that the required tasks are done. Second, the participants must have a positive attitude. This includes the willing- ness to do some of the thankless work that is required for success. Third, the participants must understand and use the fundamentals of chemical process safety in the design, construction, and operation of their plants. Fourth, everyone must learn from the experience of history or be doomed to repeat it. It is especially recommended that employees (1) read and understand

Fundamentals Attitude \ Experience

Figure 1-2 The ingredients of a successful safety program.

4 Chapter 1 Introduction

case histories of past accidents and (2) ask people in their own and other organizations for their experience and advice. Fifth, everyone should recognize that safety takes time. This includes time to study, time to do the work, time to record results (for history), time to share experiences, and time to train or be trained. Sixth, everyone (you) should take the responsibility to contribute to the safety program. A safety program must have the commitment from all levels within the organization. Safety must be given importance equal to production.

The most effective means of implementing a safety program is to make it everyone's responsibility in a chemical process plant. The older concept of identifying a few employees to be responsible for safety is inadequate by today's standards. All employees have the responsibility to be knowledgeable about safety and to practice safety.

It is important to recognize the distinction between a good and an outstanding safety program.

A good safety program identifies and eliminates existing safety hazards. An outstanding safety program has management systems that prevent the existence of safety hazards.

A good safety program eliminates the existing hazards as they are identified, whereas an outstanding safety program prevents the existence of a hazard in the first place.

The commonly used management systems directed toward eliminating the existence of hazards include safety reviews, safety audits, hazard identification techniques, checklists, and proper application of technical knowledge.

1-2 Engineering Ethics Most engineers are employed by private companies that provide wages and benefits for their services. The company earns profits for its shareholders, and engineers must provide a service to the company by maintaining and improving these profits. Engineers are responsible for minimizing losses and providing a safe and secure environment for the company's employees. En- gineers have a responsibility to themselves, fellow workers, family, community, and the engineering profession. Part of this responsibility is described in the Engineering Ethics statement developed by the American Institute of Chemical Engineers (AICHE), shown in Table 1-1.

1-3 Accident and Loss Statistics Accident and loss statistics are important measures of the effectiveness of safety programs. These statistics are valuable for determining whether a process is safe or whether a safety procedure is working effectively.

Many statistical methods are available to characterize accident and loss performance. These statistics must be used carefully. Like most statistics they are only averages and do not reflect the potential for single episodes involving substantial losses. Unfortunately, no single method is capable of measuring all required aspects. The three systems considered here are

1-3 Accident and Loss Statistics 5

Table 1-1 American Institute of Chemical Engineers Code of Professional Ethics

Fundamental principles

Engineers shall uphold and advance the integrity, honor, and dignity of the engineering profession by 1. using their knowledge and skill for the enhancement of human welfare; 2. being honest and impartial and serving with fidelity the public, their employers, and clients; 3. striving to increase the competence and prestige of the engineering profession.

Fundamental canons

1. Engineers shall hold paramount the safety, health, and welfare of the public in the performance of their professional duties.

2. Engineers shall perform services only in areas of their competence. 3. Engineers shall issue public statements only in an objective and truthful manner. 4. Engineers shall act in professional matters for each employer or client as faithful agents or trustees,

and shall avoid conflicts of interest. 5. Engineers shall build their professional reputations on the merits of their services. 6. Engineers shall act in such a manner as to uphold and enhance the honor, integrity, and dignity of the

engineering profession. 7. Engineers shall continue their professional development throughout their careers and shall provide

opportunities for the professional development of those engineers under their supervision.

OSHA incidence rate, fatal accident rate (FAR), and fatality rate, or deaths per person per year.

All three methods report the number of accidents and/or fatalities for a fixed number of workers during a specified period.

OSHA stands for the Occupational Safety andHealth Administration of the United States government. OSHA is responsible for ensuring that workers are provided with a safe working environment. Table 1-2 contains several OSHA definitions applicable to accident statistics.

The OSHA incidence rate is based on cases per 100 worker years. A worker year is assumed to contain 2000 hours (50 work weekslyear X 40 hourslweek). The OSHA incidence rate is therefore based on 200,000 hours of worker exposure to a hazard. The OSHA incidence rate is calculated from the number of occupational injuries and illnesses and the total number of employee hours worked during the applicable period. The following equation is used:

Number of injuries and OSHA incidence rate illnesses X 200,000 (based on injuries = Total hours worked by

and illness) all employees during

period covered.

Table 1-2 Glossary of Terms Used by OSHA and Industry to Represent Work-Related L o s s e ~ ~ . ~

Term Definition

First aid Any one-time treatment and any follow-up visits for the purpose of observation of minor scratches, cuts, burns, splinters, and so forth that do not ordinarily require medical care. Such one-time treatment and follow-up visits for the purpose of observation are considered first aid even though provided by a physician or registered professional personnel.

Incident rate

Lost workdays

Medical treatment

Occupational injury

Occupational illness

Number of occupational injuries and/or illnesses or lost workdays per 100 full-time employees. Number of days (consecutive or not) after but not including the day of injury or illness during which the employee would have worked but could not do so, that is, during which the employee could not perform all or any part of his or her normal assignment during all or any part of the workday or shift because of the occupational injury or illness. Treatment administered by a physician or by registered professional personnel under the standing orders of a physician. Medical treatment does not include first aid treatment even though provided by a physician or registered professional personnel. Any injury such as a cut, sprain, or burn that results from a work accident or from a single instantaneous exposure in the work environment. Any abnormal condition or disorder, other than one resulting from an occupational injury, caused by exposure to environmental factors associated with employment. It includes acute and chronic illnesses or diseases that may be caused by inhalation, absorption, ingestion, or direct contact.

Recordable cases Cases involving an occupational injury or occupational illness, including deaths.

Recordable fatality cases Injuries that result in death, regardless of the time between the injury and death or the length of the illness.

Recordable nonfatal Cases of occupational injury or illness that do not involve fatalities or lost cases without lost workdays but do result in (1) transfer to another job or termination of workdays employment or (2) medical treatment other than first aid or (3) diagnosis

of occupational illness or (4) loss of consciousness or (5) restriction of work or motion.

Recordable lost workday Injuries that result in the injured person not being able to perform their cases due to restricted regular duties but being able to perform duties consistent with their duty normal work. Recordable cases with Injuries that result in the injured person not being able to return to work days away from work on their next regular workday. Recordable medical cases Injuries that require treatment that must be administered by a physician or

under the standing orders of a physician. The injured person is able to return to work and perform his or her regular duties. Medical injuries include cuts requiring stitches, second-degree burns (burns with blisters), broken bones, injury requiring prescription medication, and injury with loss of consciousness.

'Injury Facts, 1999 ed. (Chicago: National Safety Council, 1999), p. 151. ZOSHA regulations, 29 CFR 1904.12.


An incidence rate can also be based on lost workdays instead of injuries and illnesses. For this case

Number of lost OSHA incidence rate

workdays X 200,000 (based on lost = Total hours worked by workdavs)

, , all employees during

period covered.

The definition of a lost workday is given in Table 1-2. The OSHA incidence rate provides information on all types of work-related injuries and

illnesses, including fatalities. This provides a better representation of worker accidents than systems based on fatalities alone. For instance, a plant might experience many small accidents with resulting injuries but no fatalities. On the other hand, fatality data cannot be extracted from the OSHA incidence rate without additional information.

TheFAR is usedmostly by the British chemicalindustry. This statistic is used here because there are some useful and interesting FAR data available in the open literature. The FAR reports the number of fatalities based on 1000 employees working their entire lifetime. The employees are assumed to work a total of 50 years. Thus the FAR is based on 10' working hours. The resulting equation is

Number of fatalities X 10'

FAR = Total hours worked by all

employees during period covered.

The last method considered is the fatality rate or deaths per person per year. This system is independent of the number of hours actually worked and reports only the number of fatalities expected per person per year. This approach is useful for performing calculations on the general population, where the number of exposed hours is poorly defined. The applicable equation is

Number of fatalities per year

Fatality rate = Total number of people in

applicable population.

Both the OSHA incidence rate and the FAR depend on the number of exposed hours. An employee working a ten-hour shift is at greater total risk than one working an eight-hour shift. A FAR can be converted to a fatality rate (or vice versa) if the number of exposed hours is known. The OSHA incidence rate cannot be readily converted to a FAR or fatality rate because it contains both injury and fatality information.


Table 1-3 Accident Statistics for Selected Industries

Industry

OSHA incident rate (cases involving days away from FAR

work and deaths) (deaths) 1985l 19982 19863 19904

Chemicals and allied products 0.49 0.35 4.0 1.2 Motor vehicles 1.08 6.07 1.3 0.6 Steel 1.54 1.28 8.0 Paper 2.06 0.81 Coal mining 2.22 0.26 40 7.3 Food 3.28 1.35 Construction 3.88 0.6 67 5.0 Agricultural 4.53 0.89 10 3.7 Meat products 5.27 0.96 Trucking 7.28 2.10 All manufacturing 1.68 1.2

'Accident Facts, 1985 ed. (Chicago: National Safety Council, 1985), p. 30. ZInjury Facts, 1999 ed. (Chicago: National Safety Council, 1999), p. 66. "rank P. Lees, Loss Prevention in the Process Industries (London: Butterworths, 1986), p. 177. 4Frank P. Lees, Loss Prevention in the Process Industries, 2d ed. (London: Butterworths, 1996), p. 219.

Example 1-1 A process has a reported FAR of 2. If an employee works a standard 8-hr shift 300 days per year, compute the deaths per person per year.

Solution Deaths per person per year = (8 hrlday) x (300 dayslyr) x (2 deaths/108 hr)

= 4.8 X lo-'.

Typical accident statistics for various industries are shown in Table 1-3. A FAR of 1.2 is reported in Table 1-3 for the chemical industry. Approximately half these deaths are due to ordinary industrial accidents (falling down stairs, being run over), the other half to chemical exposure^.^

The FAR figures show that if 1000 workers begin employment in the chemical industry, 2 of the workers will die as a result of their employment throughout all of their working life- times. One of these deaths will be due to direct chemical exposure. However, 20 of these same

2T. A. Kletz, "Eliminating Potential Process Hazards," Chemical Engineering (Apr. 1,1985).


Table 1-4 Fatality Statistics for Common Nonindustrial Act i~i t iesl .~

FAR Fatality rate (deaths11 0' (deaths per

Activity hours) person per year) Voluntary activity

Staying at home Traveling by

Car 57 17 X lo-" Bicycle 96 Air 240 Motorcycle 660

Canoeing 1000 Rock climbing 4000 4 x lo-5 Smoking (20 cigaretteslday) 500 X

Involuntary activity Struck by meteorite 6 X lo-" Struck by lightning (U.K.) 1 x lo-7 Fire (U.K.) 150 X lo-' Run over by vehicle 600 X lo-'

'Frank P. Lees, Loss Prevention in the Process Industries (London: Butterworths, 1986), p. 178. ZFrank P. Lees, Loss Prevention in the Process Industries, 2d ed. (London: Buttenvorths, 1996), p. 9/96.

1000 people will die as a result of nonindustrial accidents (mostly at home or on the road) and 370 will die from disease. Of those that perish from disease, 40 will die as a direct result of ~moking .~

Table 1-4 lists the FARs for various common activities. The table is divided into voluntary and involuntary risks. Based on these data, it appears that individuals are willing to take a substantially greater risk if it is voluntary. It is also evident that many common everyday activities are substantially more dangerous than working in a chemical plant.

For example, Table 1-4 indicates that canoeing is much more dangerous than traveling by motorcycle, despite general perceptions otherwise. This phenomenon is due to the number of exposed hours. Canoeing produces more fatalities per hour of activity than traveling by motorcycle. The total number of motorcycle fatalities is larger because more people travel by motorcycle than canoe.

Example 1-2 If twice as many people used motorcycles for the same average amount of time each, what will happen to (a) the OSHA incidence rate, (b) the FAR, (c) the fatality rate, and (d) the total number of fatalities?

"letz, "Eliminating Potential Process Hazards.''


Solution a. The OSHA incidence rate will remain the same. The number of injuries and deaths will

double, but the total number of hours exposed will double as well. b. The FAR will remain unchanged for the same reason as in part a. c. The fatality rate, or deaths per person per year, will double. The fatality rate does not depend

on exposed hours. d. The total number of fatalities will double.

Example 1-3 If all riders used their motorcycles twice as much, what will happen to (a) the OSHA incidence rate, (b) the FAR, (c) the fatality rate, and (d) the total number of fatalities? Solution

a. The OSHA incidence rate will remain the same. The same reasoning applies as for Example 1-2, part a.

b. The FAR will remain unchanged for the same reason as in part a. c. The fatality rate will double. Twice as many fatalities will occur within this group. d. The number of fatalities will double.

Example 1-4 A friend states that more rock climbers are killed traveling by automobile than are killed rock climbing. Is this statement supported by the accident statistics?

Solution The data from Table 1-4 show that traveling by car (FAR = 57) is safer than rock climbing (FAR = 4000). Rock climbing produces many more fatalities per exposed hour than traveling by car. How- ever, the rock climbers probably spend more time traveling by car than rock climbing. As a result, the statement might be correct but more data are required.

Recognizing that the chemical industry is safe, why is there so much concern about chemical plant safety? The concern has to do with the industry's potential for many deaths, as, for example, in the Bhopal, India, tragedy. Accident statistics do not include information on the total number of deaths from a single incident. Accident statistics can be somewhat misleading in this respect. For example, consider two separate chemical plants. Both plants have a probability of explosion and complete devastation once every 1000 years. The first plant employs a single operator. When the plant explodes, the operator is the sole fatality. The second plant employs 10 operators. When this plant explodes all 10 operators succumb. In both cases the FAR and OSHA incidence rate are the same; the second accident kills more people, but there are a correspondingly larger number of exposed hours. In both cases the risk taken by an individual operator is the same.4

It is human nature to perceive the accident with the greater loss of life as the greater tragedy. The potential for large loss of life gives the perception that the chemical industry is unsafe.

4Kletz, "Eliminating Potential Process Hazards."

1-3 Accident and Loss Statistics

N u m b e r o f Accidents Figure 1-3 The accident pyramid.

Loss data5 published for losses after 1966 and in 10-year increments indicate that the total number of losses, the total dollar amount lost, and the average amount lost per incident have steadily increased. The total loss figure has doubled every 10 years despite increased efforts by the chemical process industry to improve safety. The increases are mostly due to an expansion in the number of chemical plants, an increase in chemical plant size, and an increase in the use of more complicated and dangerous chemicals.

Property damage and loss of production must also be considered in loss prevention. These losses can be substantial. Accidents of this type are much more common than fatalities. This is demonstrated in the accident pyramid shown in Figure 1-3. The numbers provided are only ap- proximate. The exact numbers vary by industry, location, and time. "No Damage" accidents are frequently called "near misses" and provide a good opportunity for companies to determine that a problem exists and to correct it before a more serious accident occurs. It is frequently said that "the cause of an accident is visible the day before it occurs." Inspections, safety reviews and careful evaluation of near misses will identify hazardous conditions that can be corrected before real accidents occur.

Safety is good business and, like most business situations, has an optimal level of activity beyond which there are diminishing returns. As shown by K l e t ~ , ~ if initial expenditures are made on safety, plants are prevented from blowing up and experienced workers are spared. This results in increased return because of reduced loss expenditures. If safety expenditures increase, then the return increases more, but it may not be as much as before and not as much as achieved by spending money elsewhere. If safety expenditures increase further, the price of the product increases and sales diminish. Indeed, people are spared from injury (good humanity), but the cost is decreased sales. Finally, even higher safety expenditures result in uncompetitive product pricing: The company will go out of business. Each company needs to determine an appropriate level for safety expenditures. This is part of risk management.

From a technical viewpoint, excessive expenditures for safety equipment to solve single safety problems may make the system unduly complex and consequently may cause new safety

SLarge Property Damage Losses in the Hydrocarbon-Chemical Industries: A Thirty-Year Review (New York: J & H Marsh & McLennan Inc., 1998), p. 2.

6T. A. Kletz, "Eliminating Potential Process Hazards."


Table 1-5 All Accidental Deaths1

Type of death 1998 Deaths

Motor-vehicle Public nonwork 38,900 Work 2,100 Home 200

Subtotal 41,200 (43.5%) Work

Nonmotor-vehicle 3,000 Motor-vehicle 2,100

Subtotal 5,100 (5.4%) Home

Nonmotor-vehicle 28,200 Motor-vehicle 200

Subtotal 28,400 (30.0%) Public2 20,000

Subtotal 20,000 (21.1%) Total accidental deaths 92,200 3

lZnjury Facts, 1999 ed. (Chicago: National Safety Council, 1999), p. 2. 2Public accidents are any accidents other than motor-vehicle accidents that occur in the use of public facilities or premises (swimming, hunting, falling, etc.) and deaths resulting from natural disasters even if they happened in the home. 3The true total is lower than the sum of the subtotals because some accidents are in more than one category.

problems because of this complexity. This excessive expense could have a higher safety return if assigned to a different safety problem. Engineers need to also consider other alternatives when designing safety improvements.

It is also important to recognize the causes of accidental deaths, as shown in Table 1-5. Be- cause most, if not all, company safety programs are directed toward preventing injuries to employees, the programs should include off-the-job safety, especially training to prevent accidents with motor vehicles.

When organizations focus on the root causes of worker injuries, it is helpful to analyze the manner in which workplace fatalities occur (see Figure 1-4). Although the emphasis of this book is the prevention of chemical-related accidents, the data in Figure 1-4 show that safety programs need to include training to prevent injuries resulting from transportation, assaults, mechanical and chemical exposures, and fires and explosions.

1-4 Acceptable Risk We cannot eliminate risk entirely. Every chemical process has a certain amount of risk associated with it. At some point in the design stage someone needs to decide if the risks are "accept-

1-4 Acceptable Risk 13

Transportation incidents

(n = 2,630)

Assaults and violent acts (n = 960)

Contact with objects and equipment (n = 941)

Falls (n = 702)

Exposure due to harmful

substances and environments

(n = 572)

Fires and explosions (n = 205)

5 10 15 20 25 30 35 40 45 Accidents (%)

Highway

Figure 1-4 The manner in which workplace fatalities occurred in 1998. The total number of workplace fatalities was 6026. Source: News, USDL 99-208 (Washington, DC: US Department of Labor, Aug. 4, 1999).

able." That is, are the risks greater than the normal day-to-day risks taken by individuals in their nonindustrial environment? Certainly it would require a substantial effort and considerable expense to design a process with a risk comparable to being struck by lightning (see Table 1-4). Is it satisfactory to design a process with a risk comparable to the risk of sitting at home? For a single chemical process in a plant composed of several processes, this risk may be too high because the risks resulting from multiple exposures are additive.'

Worker struck by vehicle

7Modern site layouts require sufficient separation of plants within the site to minimize risks of multiple exposures.

Homicide

highway

Suicide

Aircraft

Struck by object

Other

Other

Electrocutions Other

I I I I I I I I I


28% More Good Than H a r m

- - - - - - - - - - - -

29% More Harm Than Good

- -

38% Same Amount of Good and Harm

Figure 1-5 Results from a public opinion survey asking the question "Would you say chemicals do more good than harm, more harm than good, or about the same amount of each?" Source: The Detroit News.

Engineers must make every effort to minimize risks within the economic constraints of the process. No engineer should ever design a process that he or she knows will result in certain human loss or injury, despite any statistics.

1-5 Public Perceptions The general public has great difficulty with the concept of acceptable risk. The major objection is due to the involuntary nature of acceptable risk. Chemical plant designers who specify the acceptable risk are assuming that these risks are satisfactory to the civilians living near the plant. Frequently these civilians are not aware that there is any risk at all.

The results of a public opinion survey on the hazards of chemicals are shown in Fig- ure 1-5. This survey asked the participants if they would say chemicals do more good than harm, more harm than good, or about the same amount of each. The results show an almost even three-way split, with a small margin to those who considered the good and harm to be equal.

Some naturalists suggest eliminating chemical plant hazards by "returning to nature." One alternative, for example, is to eliminate synthetic fibers produced by chemicals and use natural fibers such as cotton. As suggested by Kletz? accident statistics demonstrate that this will result in a greater number of fatalities because the FAR for agriculture is higher.

8T. A. Kletz, "Eliminating Potential Process Hazards."

1-6 The Nature of the Accident Process

Table 1-6 Three Types of Chemical Plant Accidents

Type of Probability Potential for Potential for accident of occurrence fatalities economic loss

Fire High Low Intermediate Explosion Intermediate Intermediate High Toxic release Low High Low

Example 1-5 List six different products produced by chemical engineers that are of significant benefit to mankind.

Solution Penicillin, gasoline, synthetic rubber, paper, plastic, concrete.

1-6 The Nature of the Accident Process Chemical plant accidents follow typical patterns. It is important to study these patterns in order to anticipate the types of accidents that will occur. As shown in Table 1-6, fires are the most common, followed by explosion and toxic release. With respect to fatalities, the order reverses, with toxic release having the greatest potential for fatalities.

Economic loss is consistently high for accidents involving explosions. The most damaging type of explosion is an unconfined vapor cloud explosion, where a large cloud of volatile and flammable vapor is released and dispersed throughout the plant site followed by ignition and explosion of the cloud. An analysis of the largest chemical plant accidents (based on worldwide accidents and 1998 dollars) is provided in Figure 1-6. As illustrated, vapor cloud explosions ac-

Other r 3%

Figure 1-6 Types of loss for large hydrocarbon- chemical plant accidents. Source: Large Property Damage Losses in the Hydrocarbon-Chemical Indus- tries: A Thirty-Year Review (New York: Marsh Inc., 1998), b. 2. Used by permission of Marsh Inc.


count for the largest percentage of these large losses. The "other" category of Figure 1-6 includes losses resulting from floods and windstorms.

Toxic release typically results in little damage to capital equipment. Personnel injuries, employee losses, legal compensation, and cleanup liabilities can be significant.

Figure 1-7 presents the causes of losses for the largest chemical accidents. By far the largest cause of loss in a chemical plant is due to mechanical failure. Failures of this type are usually due to a problem with maintenance. Pumps, valves, and control equipment will fail if not properly maintained. The second largest cause is operator error. For example, valves are not opened or closed in the proper sequence or reactants are not charged to a reactor in the correct order. Process upsets caused by, for example, power or cooling water failures account for 11 % of the losses.

Human error is frequently used to describe a cause of losses. Almost all accidents, except those caused by natural hazards, can be attributed to human error. For instance, mechanical failures could all be due to human error as a result of improper maintenance or inspection. The

Mechanical Operator Unknown Process Natural Design Sabotage error upsets hazards and arson

Figure 1-7 Causes of losses in the largest hydrocarbon-chemical plant accidents. Source: Large Property Damage Losses in the Hydrocarbon-Chemical Industries: A Thirty-Year Review (New York: J & H Marsh & McLennan Inc., 1998), p. 2. Used by permission of Marsh Inc.

1-6 The Nature of the Accident Process 17

Piping systems

Miscellaneous or unknown I Storage tanks

Reactor piping systems

Process holding tanks

Heat exchangers

Valves

Process towers

Compressors

Pumps

Gauges

0 5 10 15 20 25 30 35 40 45 50 Number of accidents

Figure 1-8 Hardware associated with largest losses. Source: A Thirty-Year Review of One Hundred of the Largest Property Damage Losses in the Hydrocarbon-Chemical Industries (New York: Marsh Inc., 1987). Reprinted by permission.

term "operator error," used in Figure 1-7, includes human errors made on-site that lead directly to the loss.

Figure 1-8 presents a survey of the type of hardware associated with large accidents. Pip- ing system failure represents the bulk of the accidents, followed by storage tanks and reactors. An interesting result of this study is that the most complicated mechanical components (pumps and compressors) are minimally responsible for large losses.

The loss distribution for the hydrocarbon and chemical industry over 5-year intervals is shown in Figure 1-9. The number and magnitude of the losses increase over each consecutive 10-year period for the past 30 years. This increase corresponds to the trend of building larger and more complex plants.

The lower losses in the last 5-year period, compared to the previous 5 years between 1987 and 1996, is likely the result of governmental regulations that were implemented in the United States during this time; that is, on February 24,1992, OSHA published its final rule "Process Safety Management of Highly Hazardous Chemicals." This rule became effective on May 26,


Figure 1-9 Loss distribution for onshore accidents for 5-year intervals over a 30-year period. (There were also 7 offshore accidents in this 30-year period.) Source: Large Property Damage Losses in the Hydrocarbon-Chemical Industries: A Thirty-Year Review (New York: J & H Marsh & McLennan Inc., 1998), p. 2. Used by permission of Marsh Inc.

g 2.0 .-

L

a, a a, 1.5 r, K .-

", 1.0 In 0 -

- B 0.5 P

1992. The impact of these regulations occurred in subsequent years. Other countries are adopt- ing similar regulations.

Accidents follow a three-step process. The following chemical plant accident illustrates these steps.

A worker walking across a high walkway in a process plant stumbles and falls toward the edge. To prevent the fall, he grabs a nearby valve stem. Unfortunately, the valve stem shears off and flammable liquid begins to spew out. A cloud of flammable vapor rapidly forms and is ignited by a nearby truck. The explosion and fire quickly spread to nearby equipment. The resulting fire lasts for six days until all flammable materials in the plant are consumed, and the plant is completely destroyed.

This disaster occurred in 1969y and led to an economic loss of $4,161,000. It demonstrates an important point: Even the simplest accident can result in a major catastrophe.

Most accidents follow a three-step sequence:

initiation (the event that starts the accident), propagation (the event or events that maintain or expand the accident), and termination (the event or events that stop the accident or diminish it in size).

In the example the worker tripped to initiate the accident. The accident was propagated by the shearing of the valve and the resulting explosion and growing fire. The event was terminated by consumption of all flammable materials.

-

(1.34) -

9 0 n e Hundred Largest Losses: A Thirty-Year Review of Property Lo.sses in the Hydrocarbon-Chemical Industries (Chicago: M & M Protection Consultants. 1986), p. 3.

27 Losses (1.48)

18 Losses -

(0,44) - (0.39) 17 Losses

5 Losses

(1.04)

16 Losses 9 Losses

1-6 The Nature of the Accident Process 19

Table 1-7 Defeating the Accident Process

Desired Step effect Procedure

--

Initiation Diminish Grounding and bonding Inerting Explosion proof electrical Guardrails and guards Maintenance procedures Hot work permits Human factors design Process design Awareness of dangerous properties of chemicals

Propagation Diminish Emergency material transfer Reduce inventories of flammable materials Equipment spacing and layout Nonflammable construction materials Installation of check and emergency shutoff valves

Termination Increase Firefighting equipment and procedures Relief systems Sprinkler systems Installation of check and emergency shutoff valves

Safety engineering involves eliminating the initiating step and replacing the propagation steps with termination events. Table 1-7 presents a few ways to accomplish this. In theory, accidents can be stopped by eliminating the initiating step. In practice this is not effective: It is unrealistic to expect elimination of all initiations. A much more effective approach is to work on all three areas to ensure that accidents, once initiated, do not propagate and will terminate as quickly as possible.

Example 1-6 The following accident report has been filed lo:

Failure of a threaded 1%" drain connection on a rich oil line at the base of an absorber tower in a large (1.35 MCFID) gas producing plant allowed the release of rich oil and gas at 850 psi and -40F. The resulting vapor cloud probably ignited from the ignition system of engine- driven recompressors. The 75' high X 10' diameter absorber tower eventually collapsed across the pipe rack and on two exchanger trains. Breaking pipelines added more fuel to the fire. Se- vere flame impingement on an 11,000-horsepower gas turbine-driven compressor, waste heat recovery and super-heater train resulted in its near total destruction.

Identify the initiation, propagation, and termination steps for this accident.

l0One Hundred Largest Losses, p. 10.

20 Chapter 1 introduction

Solution Initiation: Failure of threaded 172" drain connection Propagation: Release of rich oil and gas, formation of vapor cloud, ignition of vapor cloud by re-

compressors, collapse of absorber tower across pipe rack Termination: Consumption of combustible materials in process

As mentioned previously, the study of case histories is an especially important step in the process of accident prevention. To understand these histories, it is helpful to know the definitions of terms that are commonly used in the descriptions (see Table 1-8).

1-7 Inherent Safety An inherently safe plant11J2 relies on chemistry and physics to prevent accidents rather than on control systems, interlocks, redundancy, and special operating procedures to prevent accidents. Inherently safer plants are tolerant of errors and are often the most cost effective. A process that does not require complex safety interlocks and elaborate procedures is simpler, eas- ier to operate, and more reliable. Smaller equipment, operated at less severe temperatures and pressures, has lower capital and operating costs.

In general, the safety of a process relies on multiple layers of protection. The first layer of protection is the process design features. Subsequent layers include control systems, interlocks, safety shutdown systems, protective systems, alarms, and emergency response plans. In- herent safety is a part of all layers of protection; however, it is especially directed toward process design features. The best approach to prevent accidents is to add process design features to prevent hazardous situations. An inherently safer plant is more tolerant of operator errors and abnormal conditions.

Although a process or plant can be modified to increase inherent safety at any time in its life cycle, the potential for major improvements is the greatest at the earliest stages of process development. At these early stages process engineers and chemists have the maximum degree of freedom in the plant and process specifications, and they are free to consider basic process alternatives, such as changes to the fundamental chemistry and technology.

The major approach to inherently safer process designs is divided into the following categories:

intensification substitution attenuation limitation of effects simplification/error tolerance

llCCPS, Guidelines for Engineering Design for Process Safety (New York: American Institute of Chem- ical Engineers, 1993).

12CCPS, Inherently Safer Chemical Processes: A Life Cycle Approach (New York: American Institute of Chemical Engineers, 1996).

1-7 Inherent Safety 21

Table 1-8 Definitions for Case Histories1

Term Definition

Accident The occurrence of a sequence of events that produce unintended injury, death, or property damage. "Accident" refers to the event, not the result of the event.

Hazard

Incident

A chemical or physical condition that has the potential for causing damage to people, property, or the environment. The loss of containment of material or energy; not all events propagate into incidents; not all incidents propagate into accidents.

Consequence A measure of the expected effects of the results of an incident. Likelihood A measure of the expected probability or frequency of occurrence of an event.

This may be expressed as a frequency, a probability of occurrence during some time interval, or a conditional probability.

Risk A measure of human injury, environmental damage, or economic loss in terms of both the incident likelihood and the magnitude of the loss or injury.

Risk analysis The development of a quantitative estimate of risk based on an engineering evaluation and mathematical techniques for combining estimates of incident conse- quences and frequencies.

Risk assessment The process by which the results of a risk analysis are used to make decisions, either through a relative ranking of risk reduction strategies or through compari- son with risk targets.

Scenario A description of the events that result in an accident or incident. The description should contain information relevant to defining the root causes.

'CCPS, Guidelines for Consequence Analysis of Chemical Releases (New York: American Institute of Chemical Engi- neers, 1999).

These five categories are the predominant ones used since the development of this concept. Some companies add or subtract categories to their program to fine-tune their understanding and application. In an attempt to make these categories more understandable, the following four words have recently been recommended to describe inherent safety:

minimize (intensification) substitute (substitution) moderate (attenuation and limitation of effects) simplify (simplification and error tolerance).

The types of inherent safety techniques that are used in the chemical industry are illustrated in Table 1-9 and are described more fully in what follows.

Minimizing entails reducing the hazards by using smaller quantities of hazardous substances in the reactors, distillation columns, storage vessels, and pipelines. When possible, hazardous materials should be produced and consumed in situ. This minimizes the storage and transportation of hazardous raw materials and intermediates.

Chapter 1 Introduction

Table 1-9 Inherent Safety Techniques

TY pe Typical techniques

Minimize (intensification)

Substitute (substitution)

Moderate (attenuation and limitation of effects)

Simplify (simplification and error tolerance)

Change from large batch reactor to a smaller continuous reactor Reduce storage inventory of raw materials Improve control to reduce inventory of hazardous intermediate chemicals Reduce process hold-up Use mechanical pump seals vs. packing Use welded pipe vs. flanged Use solvents that are less toxic Use mechanical gauges vs. mercury Use chemicals with higher flash points, boiling points, and other less hazardous

properties Use water as a heat transfer fluid instead of hot oil Use vacuum to reduce boiling point Reduce process temperatures and pressures Refrigerate storage vessels Dissolve hazardous material in safe solvent Operate at conditions where reactor runaway is not possible Place control rooms away from operations Separate pump rooms from other rooms Acoustically insulate noisy lines and equipment Barricade control rooms and tanks Keep piping systems neat and visually easy to follow Design control panels that are easy to comprehend Design plants for easy and safe maintenance Pick equipment that requires less maintenance Pick equipment with low failure rates Add fire- and explosion-resistant barricades Separate systems and controls into blocks that arc easy to comprehend and

understand Label pipes for easy "walking the line" Label vessels and controls to enhance understanding

Vapor released from spills can be minimized by designing dikes so that flammable and toxic materials will not accumulate around leaking tanks. Smaller tanks also reduce the hazards of a release.

While minimization possibilities are being investigated, substitutions should also be considered as an alternative or companion concept; that is, safer materials should be used in place of hazardous ones. This can be accomplished by using alternative chemistry that allows the use of less hazardous materials or less severe processing conditions. When possible, toxic or flammable solvents should be replaced with less hazardous solvents (for example, water-based paints and adhesives and aqueous or dry flowable formulations for agricultural chemicals).

Another alternative to substitution is moderation, that is, using a hazardous material un-

1-8 Four Significant Disasters 23

der less hazardous conditions. Less hazardous conditions or less hazardous forms of a material include (1) diluting to a lower vapor pressure to reduce the release concentration, (2) refriger- ating to lower the vapor pressure, (3) handling larger particle size solids to minimize dust, and (4) processing under less severe temperature or pressure conditions.

Containment buildings are sometimes used to moderate the impact of a spill of an especially toxic material. When containment is used, special precautions are included to ensure worker protection, such as remote controls, continuous monitoring, and restricted access.

Simpler plants are friendlier than complex plants because they provide fewer opportunities for error and because they contain less equipment that can cause problems. Often, the reason for complexity in a plant is the need to add equipment and automation to control the hazards. Simplification reduces the opportunities for errors and misoperation. For example, (1) piping systems can be designed to minimize leaks or failures, (2) transfer systems can be designed to minimize the potential for leaks, (3) process steps and units can be separated to prevent the domino effect, (4) fail-safe valves can be added, (5) equipment and controls can be placed in a logical order, and (6) the status of the process can be made visible and clear at all times.

The design of an inherently safe and simple piping system includes minimizing the use of sight glasses, flexible connectors, and bellows, using welded pipes for flammable and toxic chemicals and avoiding the use of threaded pipe, using spiral wound gaskets and flexible graphite- type gaskets that are less prone to catastrophic failures, and using proper support of lines to minimize stress and subsequent failures.

1-8 Four Significant Disasters The study of case histories provides valuable information to chemical engineers involved with safety. This information is used to improve procedures to prevent similar accidents in the future.

The four most cited accidents (Flixborough, England; Bhopal, India; Seveso, Italy; and Pasadena, Texas) are presented here. All these accidents had a significant impact on public perceptions and the chemical engineering profession that added new emphasis and standards in the practice of safety. Chapter 13 presents case histories in considerably more detail.

The Flixborough accident is perhaps the most documented chemical plant disaster. The British government insisted on an extensive investigation.

Flixborough, England The accident at Flixborough, England, occurred on a Saturday in June 1974. Although it

was not reported to any great extent in the United States, it had a major impact on chemical engineering in the United Kingdom. As a result of the accident, safety achieved a much higher priority in that country.

The Flixborough Works of Nypro Limited was designed to produce 70,000 tons per year of caprolactam, a basic raw material for the production of nylon. The process uses cyclohexane,


which has properties similar to gasoline. Under the process conditions in use at Flixborough (155C and 7.9 atm), the cyclohexane volatilizes immediately when depressurized to atmo- spheric conditions.

The process where the accident occurred consisted of six reactors in series. In these reactors cyclohexane was oxidized to cyclohexanone and then to cyclohexanol using injected air in the presence of a catalyst. The liquid reaction mass was gravity-fed through the series of reactors. Each reactor normally contained about 20 tons of cyclohexane.

Several months before the accident occurred, reactor 5 in the series was found to be leaking. Inspection showed a vertical crack in its stainless steel structure. The decision was made to remove the reactor for repairs. An additional decision was made to continue operating by connecting reactor 4 directly to reactor 6 in the series. The loss of the reactor would reduce the yield but would enable continued production because unreacted cyclohexane is separated and recycled at a later stage.

The feed pipes connecting the reactors were 28 inches in diameter. Because only 20-inch pipe stock was available at the plant, the connections to reactor 4 and reactor 6 were made using flexible bellows-type piping, as shown in Figure 1-10. It is hypothesized that the bypass pipe section ruptured because of inadequate support and overflexing of the pipe section as a result of internal reactor pressures. Upon rupture of the bypass, an estimated 30 tons of cyclohexane volatilized and formed a large vapor cloud. The cloud was ignited by an unknown source an estimated 45 seconds after the release.

The resulting explosion leveled the entire plant facility, including the administrative offices. Twenty-eight people died, and 36 others were injured. Eighteen of these fatalities occurred in the main control room when the ceiling collapsed. Loss of life would have been substantially greater had the accident occurred on a weekday when the administrative offices were filled with employees. Damage extended to 1821 nearby houses and 167 shops and factories. Fifty-three civilians were reported injured. The resulting fire in the plant burned for over 10 days.

This accident could have been prevented by following proper safety procedures. First, the

Temporary Pipe Sect ion

Be l lows

Figure 1-10 A failure of a temporary pipe section replacing reactor 5 caused the Flixborough accident.


bypass line was installed without a safety review or adequate supervision by experienced engineering personnel. The bypass was sketched on the floor of the machine shop using chalk! Sec- ond, the plant site contained excessively large inventories of dangerous compounds. This included 330,000 gallons of cyclohexane, 66,000 gallons of naphtha, 11,000 gallons of toluene, 26,400 gallons of benzene, and 450 gallons of gasoline. These inventories contributed to the fires after the initial blast. Finally, the bypass modification was substandard in design. As a rule, any modifications should be of the same quality as the construction of the remainder of the plant.

Bhopal, India The Bhopal, India, accident, on December 3, 1984, has received considerably more at-

tention than the Flixborough accident. This is due to the more than 2000 civilian casualties that resulted.

The Bhopal plant is in the state of Madhya Pradesh in central India. The plant was partially owned by Union Carbide and partially owned locally.

The nearest civilian inhabitants were 1.5 miles away when the plant was constructed. Be- cause the plant was the dominant source of employment in the area, a shantytown eventually grew around the immediate area.

The plant pesticides. An intermediate compound in this process is methyl isocyanate (MIC). MIC is an extremely dangerous compound. It is reactive, toxic, volatile, and flammable. The maximum exposure concentration of MIC for workers over an 8-hour period is 0.02 ppm (parts per million). Individuals exposed to concentrations of MIC vapors above 21 ppm experience severe irritation of the nose and throat. Death at large concentrations of vapor is due to respiratory distress.

MIC demonstrates a number of dangerous physical properties. Its boiling point at atmo- spheric conditions is 39.1C, and it has a vapor pressure of 348 mm Hg at 20C. The vapor is about twice as heavy as air, ensuring that the vapors will stay close to the ground once released.

MIC reacts exothermically with water. Although the reaction rate is slow, with inadequate cooling the temperature will increase and the MIC will boil. MIC storage tanks are typically re- frigerated to prevent this problem.

The unit using the MIC was not operating because of a local labor dispute. Somehow a storage tank containing a large amount of MIC became contaminated with water or some other substance. A chemical reaction heated the MIC to a temperature past its boiling point. The MIC vapors traveled through a pressure relief system and into a scrubber and flare system installed to consume the MIC in the event of a release. Unfortunately, the scrubber and flare systems were not operating, for a variety of reasons. An estimated 25 tons of toxic MIC vapor was released. The toxic cloud spread to the adjacent town, killing over 2000 civilians and injuring an estimated 20,000 more. No plant workers were injured or killed. No plant equipment was damaged.

The exact cause of the contamination of the MIC is not known. If the accident was caused by a problem with the process, a well-executed safety review could have identified the problem. The scrubber and flare system should have been fully operational to prevent the release. Inven- tories of dangerous chemicals, particularly intermediates, should also have been minimized.


Methyl isocyanate route

CH3NH2 + COCI, - CH3N = C = 0 + 2HCI Methylamine Phosgene Methyl isocyanate

0 II

OH 0 - CNHCH3 I I

CH3N=C=0 +

a-Naphthol Carbaryl

Nonmethyl isocyanate route

a-Na~hthol chloroformate

Figure 1-1 1 The upper reaction is the methyl isocyanate route used at Bhopal. The lower reaction suggests an alternative reaction scheme using a less hazardous intermediate. Adapted from Chemical and Engineering News (Feb. 1 1 , 1985), p. 30.

The reaction scheme used at Bhopal is shown at the top of Figure 1-1 1 and includes the dangerous intermediate MIC. An alternative reaction scheme is shown at the bottom of the figure and involves a less dangerous chloroformate intermediate. Another solution is to redesign the process to reduce the inventory of hazardous MIC. One such design produces and con- sumes the MIC in a highly localized area of the process, with an inventory of MIC of less than 20 pounds.

Seveso, Italy Seveso is a small town of approximately 17,000 inhabitants, 15 miles from Milan, Italy.

The plant was owned by the Icmesa Chemical Company. The product was hexachlorophene, a bactericide, with trichlorophenol produced as an intermediate. During normal operation, a


small amount of TCDD (2,3,7,8-tetrachlorodibenzoparadioxin) is produced in the reactor as an undesirable side-product.

TCDD is perhaps the most potent toxin known to humans. Animal studies have shown TCDD to be fatal in doses as small as lo-' times the body weight. Because TCDD is also in- soluble in water, decontamination is difficult. Nonlethal doses of TCDD result in chloracne, an acne-like disease that can persist for several years.

On July 10, 1976, the trichlorophenol reactor went out of control, resulting in a higher than normal operating temperature and increased production of TCDD. An estimated 2 kg of TCDD was released through a relief system in a white cloud over Seveso. A subsequent heavy rain washed the TCDD into the soil. Approximately 10 square miles were contaminated.

Because of poor communications with local authorities, civilian evacuation was not started until several days later. By then, over 250 cases of chloracne were reported. Over 600 people were evacuated, and an additional 2000 people were given blood tests. The most se- verely contaminated area immediately adjacent to the plant was fenced, the condition it re- mains in today.

TCDD is so toxic and persistent that for a smaller but similar release of TCDD in Du- phar, India, in 1963 the plant was finally disassembled brick by brick, encased in concrete and dumped into the ocean. Less than 200 g of TCDD was released, and the contamination was confined to the plant. Of the 50 men assigned to clean up the release, 4 eventually died from the exposure.

The Seveso and Duphar accidents could have been avoided if proper containment systems had been used to contain the reactor releases. The proper application of fundamental engineering safety principles would have prevented the two accidents. First, by following proper procedures, the initiation steps would not have occurred. Second, by using proper hazard evaluation procedures, the hazards could have been identified and corrected before the accidents occurred.

Pasadena, Texas A massive explosion in Pasadena, Texas, on October 23, 1989, resulted in 23 fatalities

Documents

Chemical Process Safety Fundamentals With Applications,2ND ED